A Classification and Identification Technology of TLS Encrypted Traffic Applications

In the paper, regarding the transport layer security encrypted traffic of the Internet, the application program identification and application protocol classification technology of encrypted traffic based on hidden Markov model and long short-term memory networks are studied, and the encrypted traffic detection prototype is constructed. The experimental results show that in terms of web application, real-time transport protocol and other encrypted traffics, the prototype can make accurate classification and identification, and the accuracy rate is generally higher than 96.5%, the recall rate higher than 92.0%, and the classification and identification time is less than 0.2 seconds, whereby the proposed method can improve the classification accuracy and real-time detection performance.